Supercompute Operations Engineering (AI)

Location: Southaven, Mississippi; Memphis, Tennessee

Company

xAI is building a super-intelligence stack spanning Colossus supercomputers, Grok models, and applications including Grok Bot, Grok Build, Grok Imagine, Grok Voice, and X.

What you will do

• Own monitoring architecture, alerting, signal quality, and monitoring automation across GPU, host, facility, power, cooling, and data center telemetry.
• Participate in 24/7 end-to-end incident response, mitigation, and control authority from customer impact through facility operations.
• Operate and sustain live data halls, resolve rack-level compute and facility infrastructure issues, and coordinate daily issue disposition across Site Operations and Facility Operations.
• Lead acceptance testing and deployment campaigns to bring new data halls fully online.
• Develop production automation for data hall bring-up, alerting, and automated incident response, and maintain operational procedures and live data hall state documentation.
• Lead cross-functional reliability work across compute, InfiniBand/RoCE networking, storage, power, cooling, hardware, and integrated operations systems, including systemic root-cause analysis and corrective actions.

Requirements

• Bachelor’s degree in Mechanical, Electrical, or Computer Engineering or a related scientific discipline and 2+ years of experience in data center, critical facilities, power plant, or large-scale compute operations; alternatively, 6+ years of equivalent experience in those environments.

Nice to have

• Hands-on experience with Prometheus/Grafana, NVIDIA DCGM, or industrial telemetry and controls including SCADA, BMS/BAS, EPMS, DCIM, PLCs, Modbus, or BACnet.
• Experience in data center site reliability, systems engineering, plant operations, or large-scale production operations.
• Experience running acceptance test campaigns for integrated products.
• Experience with AI/ML infrastructure or supercomputing environments, including GB200/NVL72 or H100/H200-class GPU fleets, NCCL, InfiniBand/RoCE, Slurm, or Kubernetes.
• Experience with fleet firmware and BMC management, including Redfish, IPMI, firmware release validation, and staged rollouts.
• Familiarity with data center hardware and plant signals covering servers, GPUs, networking, power, cooling, and software telemetry.
• Large-scale incident command experience and calm technical leadership on an SRE/NOC bridge or plant control room.
• Experience writing and operating playbooks, runbooks, standard operating procedures, methods of procedure, or emergency operations procedures.
• Proficiency in Python and Bash scripting for automation and analysis, plus general experience with a systems language such as C, C++, Java, Go, or Rust.
• Ability to collaborate with cross-functional, cross-country, and multi-disciplinary teams.

Culture & Benefits

• Work directly with Site Operations and Facility Operations technicians in data halls and the field.
• Collaborate across compute, network, storage, facility, and operations disciplines.
• Work in a 24/7 operational environment focused on GPU uptime, safe execution, and continuous response improvement.